Cells gradually matured, retaining blood sugar control for months
For generations, type 1 diabetes has represented a particular cruelty of biology — the body erasing the very cells it needs to survive. Researchers at Karolinska Institutet and KTH Royal Institute of Technology in Sweden have now brought the long-imagined remedy meaningfully closer, developing a method to grow mature, functional insulin-producing cells from human stem cells that successfully restored blood sugar regulation in diabetic mice for months. The advance addresses not one but several compounding failures that have frustrated this field — immaturity, impurity, and inconsistency — suggesting that the gap between laboratory promise and clinical reality may finally be narrowing.
- Decades of attempts to grow replacement insulin cells in the lab have repeatedly stumbled on the same obstacles: cells that were too immature, too impure, or too unpredictable to be trusted in a living body.
- The Swedish team's pivotal insight was architectural — allowing cells to form natural three-dimensional clusters rather than flat layers, a shift that dramatically reduced unwanted cell types and sharpened glucose responsiveness.
- Transplanted into the transparent anterior chamber of diabetic mice's eyes — a site chosen precisely because it allows live observation without invasive surgery — the cells matured in place and held blood sugar in check for several months.
- The research directly confronts the trio of problems that have stalled stem cell therapy for type 1 diabetes: cell maturity, batch purity, and reproducibility across experiments.
- Clinical trials of stem cell therapies are already underway, and this breakthrough raises the prospect of patient-specific treatments grown from a person's own cells, potentially sidestepping the immune rejection that has shadowed earlier approaches.
Type 1 diabetes is a disease of absence — the immune system destroys the pancreatic cells that produce insulin, leaving the body unable to regulate blood sugar without constant external intervention. The theoretical remedy has long been clear: grow replacement insulin cells in the lab and transplant them back. The execution, however, has proven stubbornly difficult. Earlier attempts yielded cells that were immature, inconsistent, or contaminated with unwanted types that risked triggering further immune complications.
A team led by Per-Olof Berggren at Karolinska Institutet, in collaboration with KTH Royal Institute of Technology, has now developed a method that appears to resolve these compounding failures at once. The critical change was rethinking how the cells are grown — shifting from flat, two-dimensional layers to natural three-dimensional clusters. This structural shift produced cells that were purer, more mature, and far more responsive to changes in glucose levels, behaving in the laboratory much as genuine pancreatic cells should.
The transplantation experiment was as inventive as the culture method. Rather than placing the cells in conventional sites, the researchers implanted them in the anterior chamber of diabetic mice's eyes — transparent enough to allow continuous, non-invasive observation. There, the cells continued maturing after transplantation and maintained blood sugar regulation for several months, offering the kind of sustained, observable performance that earlier approaches could not reliably deliver.
Researchers including Siqin Wu of Spiber Technologies AB and professor Fredrik Lanner pointed toward the larger ambition: not merely a proof of concept in mice, but a pathway to clinical use. The ultimate vision is patient-specific therapy, where cells derived from a person's own stem cells could be returned to their body without provoking immune rejection. Several clinical trials are already in progress, and this work — backed by the Swedish Research Council, the Knut and Alice Wallenberg Foundation, the Novo Nordisk Foundation, and the European Research Council — may help clear the obstacles those trials have encountered. Larger animal studies and, eventually, human trials will determine whether the promise holds.
Type 1 diabetes is a disease of absence. The immune system, in a case of cellular mistaken identity, destroys the pancreas's insulin-producing cells—the ones responsible for pulling glucose out of the bloodstream and into the body's tissues. Without them, blood sugar spirals into dangerous territory. For decades, researchers have imagined a straightforward fix: grow new insulin cells in the lab and put them back. The problem has always been execution. Earlier attempts produced cells that were immature, inconsistent, or contaminated with unwanted cell types. Now, a team of Swedish scientists says they have solved that problem.
Per-Olof Berggren and his colleagues at Karolinska Institutet, working with researchers at KTH Royal Institute of Technology, developed a refined method for coaxing human stem cells into becoming insulin-producing cells—and crucially, making them do it reliably. The key was rethinking the culture process itself. Instead of growing the cells in flat layers, the researchers allowed them to form natural three-dimensional clusters. This simple shift had outsized effects: fewer unwanted cell types emerged, and the cells that did develop showed a much stronger ability to sense and respond to glucose.
In the laboratory, these newly created cells behaved like the real thing. They released insulin when glucose levels rose. They demonstrated the kind of responsiveness that functional pancreatic cells should have. But the real test came when the team transplanted them into diabetic mice. Rather than implanting them into the abdomen or elsewhere, the researchers placed the cells in the anterior chamber of the eye—a choice that might sound unusual but served a purpose. The eye is transparent and accessible, allowing scientists to watch the cells develop and function over time without invasive surgery. What they observed was encouraging: the cells gradually matured after transplantation and maintained their ability to regulate blood sugar for several months.
This matters because it addresses a cluster of problems that have stalled stem cell therapy for type 1 diabetes. One fundamental issue has been maturity. Lab-grown cells often lack the sophistication of cells that have developed naturally in the body. Another has been purity—the presence of unwanted cell types that could trigger immune responses or cause other complications. A third has been consistency. Earlier protocols produced variable results, making it difficult to predict whether a batch of cells would work. The Swedish team's approach appears to tackle all three at once.
Siqin Wu, a researcher at Spiber Technologies AB and co-author of the work, and Fredrik Lanner, a professor at Karolinska Institutet, both emphasized the potential for what comes next. The goal is not merely to prove the concept works in mice—it is to move toward clinical translation, toward actual patients. The long-term vision is patient-specific cell therapy, where cells grown from a person's own stem cells could be transplanted back without triggering immune rejection. Several such trials are already underway, but they have been hampered by the very challenges this research appears to have overcome.
The work was supported by a constellation of Swedish and European funding bodies, including the Swedish Research Council, the Knut and Alice Wallenberg Foundation, the Novo Nordisk Foundation, and the European Research Council. Some of the researchers involved have ties to biotech companies and have filed patent applications related to the work—a detail worth noting, though not unusual in translational research. The next phase will be to test whether this approach works as well in larger animals and, eventually, whether it can be safely and effectively deployed in human patients. For people living with type 1 diabetes, the promise is straightforward: a way to restore the cells their immune system destroyed, without the need for daily insulin injections or the constant vigilance that blood sugar management demands.
Notable Quotes
We have developed a method that reliably produces high-quality insulin-producing cells from multiple human stem cell lines, opening opportunities for patient-specific cell therapies that could reduce immune rejection.— Per-Olof Berggren, Karolinska Institutet
This could solve several problems that have previously hindered stem cell-based treatments for type 1 diabetes, with the goal of moving toward clinical translation.— Fredrik Lanner, Karolinska Institutet
The Hearth Conversation Another angle on the story
Why does it matter that these cells were grown in three-dimensional clusters instead of flat layers?
Because shape changes behavior. In flat cultures, cells don't interact with each other the way they do in the body. Three-dimensional clusters let them communicate, mature, and specialize more naturally. It's the difference between studying a single neuron in isolation and watching a whole network light up.
The researchers put the cells in the eye. That seems like an odd choice for treating diabetes.
It is, but it's brilliant for observation. The eye is transparent, so they could watch the cells develop without cutting the mouse open repeatedly. It's a window into what's happening. Once they prove the concept works, they'll move to more practical locations.
What's the immune rejection problem they keep mentioning?
When you transplant cells from one person into another, the immune system often attacks them as foreign invaders. But if you grow cells from a patient's own stem cells and put them back, the body recognizes them as self. That's the dream—personalized medicine that sidesteps rejection entirely.
How long have scientists been trying to do this?
Decades. The idea of replacing lost insulin cells has been obvious for a long time. The hard part was making it work reliably. Earlier attempts produced inconsistent results—some batches worked, others didn't. This method appears to have cracked that consistency problem.
What happens next?
Larger animal studies, probably. Then human trials. The researchers are explicit about aiming for clinical translation. But there's always a gap between what works in mice and what works in people. That's the real test ahead.